LA-U R- 0 2 -57 5-6 Approved for public release; distribution is unlimited.

Title:

Author(s):

Submitted to:

Los Alamos NATIONAL L A B 0 RAT0 RY

Software Agent Technology in the Laboratory

Torsten A. Staab ESA-AET, Automation & Robotics Team Los Alamos National Laboratory, Los Alamos, NM, USA

LabMation'2002 Conference Philadelphia, PA, USA October 2002

Los Alamos National Laboratory, an affirmative actiodequal opportunity employer, is operated by the University of California for the U.S. Department of Energy under contract W-7405-ENG-36. By acceptance of this article, the publisher recognizes that the US. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or to allow others to do so, for US. Government purposes. Los Alamos National Laboratory requests that the publisher identify this article as work performed under the auspices of the US. Department of Energy. Los Alamos National Laboratory strongly supports academic freedom and a researcher's right to publish; as an institution, however, the Laboratory does not endorse the viewpoint of a publication or guarantee its technical correctness.

Form 836 (10/96)

To be submitted to

LabMationZOO2 Conference Philadelphia, PA, USA October 2002

Title: Software Agent Technology in the Laboratory

Author: Torsten Staab Los Alamos National Laboratory ESA-AET, Automation & Robotics Team Mailstop 5580 Los Alamos, NM 87545 Phone: (505) 665-7345 E-Mail: tstaab@lanl.gov

Abstract

The IT (Information Technology) environment in today's laboratories is characterized as being highly distributed, heterogeneous, and in some instances extremely dynamic. Larger organizations have to deal with hundreds of different systems, ranging from stand- alone workstations and devices in laboratories to fully integrated LIMS (Laboratory Information Management System) and ERP (Enterprise Resource Planning) systems. An information system operating in such an environment must handle several emerging problems, such as heterogeneous hardware and software platforms, as well as distributed information sources and capabilities. It is also expected that the IT infrastructure scales well, easily integrates with legacy systems, allows resource sharing, and supports day-to- day operations such as information retrieval, data storage, validation, tracking, replication, and archival in a fully automated fashion.

By using real-world examples, this presentation will illustrate how software agent technology can be used to manage the ever increasing IT complexity and user demands in the laboratory of the future.

To be submitted to

LabMationZOO2 Conference Philadelphia, PA, USA October 2002

Title: Software Agent Technology in the Laboratory

Author: Torsten Staab Los Alamos National Laboratory ESA-AET, Automation & Robotics Team Mailstop J580 Los Alamos, NM 87545 Phone: (505) 665-7345 E-Mail: tstaab@lanl.gov

Abstract

The IT (Information Technology) environment in today's laboratories is characterized as being highly distributed, heterogeneous, and in some instances extremely dynamic. Larger organizations have to deal with hundreds of different systems, ranging from stand- alone workstations and devices in laboratories to fully integrated LIMS (Laboratory Information Management System) and ERP (Enterprise Resource Planning) systems. An information system operating in such an environment must handle several emerging problems, such as heterogeneous hardware and software platforms, as well as distributed information sources and capabilities. It is also expected that the IT infrastructure scales well, easily integrates with legacy systems, allows resource sharing, and supports day-to- day operations such as information retrieval, data storage, validation, tracking, replication, and archival in a fully automated fashion.

By using real-world examples, this presentation will illustrate how software agent technology can be used to manage the ever increasing IT complexity and user demands in the laboratory of the future.

T

Software Agent Technology the Laboratory

in

Torsten A. Staab

LabMation 2002 Conference Philadelphia, PA, USA

October lst, 2002 f f

-4 Overview

/

-@ Software Agents Origins & Definitions

Agent Technology Applications & Users

Agent Typology

@b Requirements for Agent-based Systems

Agent Architecture & Languages

Collaborative Agent Technology in the Lab

Issues with Agent Technologies

Conclusions - 2 -

4 Origins of Software Agent

Software agent technology can be traced back to Hewitt's Concurrent Actor Model (1977)

Actor: a self-contained, concurrently executing object

with encapsulated internal state, capable of

responding to messages from other similar objects

- 3 -

A Definitions I

@> Agent: One that is authorized to act for another. Agents possess the characteristics of delegacy, competency, and amenabili&

Examples of human agents: booking agents, sales agents, politicians.

Software Agent: An artificial agent that operates in a software environment.

Intelligent Software Agent: A software agent that uses Artificial Intelligence (AI) in the pursuit of the goals of its clients.

- 4 -

A Agent Technology Applications & Users

Workflow Management, Network Management, Air-traffic Control,

Business Process Re-engineering, Data Mining, Information

RetrievaI/Management, Electronic Commerce, Education,

Personal Digital Assistants (PDAs), E-mail, Digital Libraries,

Command and Control, Smart Databases, Scheduling ...

e Users:

Alcatel, Apple, AT&T, BT, Daimler-Benz, DEC, HP, IBM, Lotus,

Microsoft, Oracle, Sharp, Reuters, Dow Jones, . . ,

- 5 -

Agent Typology

- 6 -

Requirements for Agent-based Systems

There are several levels at which agent-based systems must agree, at least in their interfaces, in order to successfully interoperate. These levels include the following:

8 Transport:

H how agents send and receive messages

@ Language:

what the individual messages mean (syntax + semantics)

@ Policy:

H how agents structure conversations

gi, Architecture:

how to connect systems in accordance with constituent

pro toco I s

- 7 -

/I Agent Architecture

t Other Agents I I I I I I I I I I I I

1 I I I I I 1 I I I I I I I I I I I I

I I I I I I I I I I I I I I I

1 Messages

Interact 1 L 1 Goals

I I I I 1 1 I

I

I I

I I I I I I I I I I

I Actions I I I I I I I I I I

I I

*Virtual Knowledge Base (VKB) 4 - 8 - 3 LosAlamos

-4 Agent Communication Languages (ACL) 112

4+ KQML (Knowledge Query & Manipulation Language) is both a message format and a message-handling protocol to support run-time knowledge sharing among agents.

most widely used ACL

based on Speech Act Theory - Illocution (Purpose of Message) - Context of Message - Content of Message

KQML messages are called Perfurmafives KQML Example: Sender: Mettle r-To1 edo AX504 Receiver: LECIS Controller In Reply To: ID9100.5 Ontology: BL.SW Language: Prolog Content: “Weight(S1234, 20, mg)”

A I . y C f V A 4 - 9 -

f l Agent Communication Languages (ACL) 212

<4@ KQML Performatives

achieve advertise as k-a bout ask-all ask-if ask-one break broadcast bro ker-a I I bro ker-one deny delete delete-all delete-one discard eos error

A J.Vl"$$ fVA *.r

eva I uate forward strea m-a I I generator subscribe insert tell monitor transport-address next unregister Pipe untell ready recommend-all recommend-one recruit-a II recruit-one register reply rest sorry standby

stream -a bout

- IO -

Collaborative Agent Technology-Why?

I 4+ Information available can be unorganized, multi-modeled, and distributed

Number and variety of data sources and services is dramatically increasing and constantly changing

I I

4k Same piece of information can be accessible from a variety of different information sources

4b Information can be ambiguous, erroneous, or obsolete

@ Enhance modularity (which reduces complexity), speed (due to parallelism), reliability (due to redundancy), flexibility

- 11 -

CoI la borative Agent Tech nology-Architectu re

I I I Info Agent p-q Info Agent 1

1 Information Layer

- 1 2 -

Collaborative Agent Technology-Implementatioi +------

1 1 Example 1: Agent-based Information Retrieval I

DB Info Agent

f t \

Web Info Agent

t

1

Collaborative Agent Technology-Implementatioi i , /

1 I Example 2 : Device Control

. . .

Tecan FE500 Li

c Network

I S iequence Detector

LIMS

Beckman ORCA Plate Handling Robot I

Device Contro I ler & Scheduler -3 P

- 14 -

I

A Collaborative Agent Technology-Irn plernentatiob I ! SOAP-based (Simple Object Access Protocol) Agents

I 1 I

What is SOAP? : I platform independent XML-based (extensible

Markup Language) communication protocol

0 transport protocol independent (runs on top of HlTP, SMTP, ...)

created by Ariba Inc., Compaq, HP, IBM, IONA, Lotus, Microsoft, SAP AG, Userland, and others

not intended to replace CORBA, DCOM, RMI

- 15 -

Collaborative Agent Technology-Implementatioi

SOAP -ENV: encodingStyle=llhttp: //schemas .xmlsoap.org/soap/encoding/ll>

I 1 Example of a SOAP Agent Request Message:

I

1

</SOAP-ENV:Header>

< SOAP - ENV : BO dy >

</SOAP-ENV:BOdy>

</SOAP-ENV:Envelope>

- 16 -

Collaborative Agent Technology-Implementatioi

Sequence Detector

Beckman ORCI

LIMS m

Device Controller & Scheduled!-

1

- 17 -

A Collaborative Agent Technology-Implementatiod

Compound Inventory 1 Information Agent I

A-

B - 18 -

Images courtesy of K-Team S.A. (Koala Robot) and Remp AG (Plate Storage System), Switzerland

Issues w/ Agent Technologies 7

1 c* security I

I 4b Responsibilities

l

Ownership

@ Legal Issues

:4P Ethical Issues

- 19-

/i Conclusions

9- is domain & application-independent

Agent Technology.. .

@ scales well

@ integrates easily with legacy systems

can be implemented using standard protocols (RMI/RPC, SOAP,

HlTP, SMTP, ...), middleware (CORBA, COM+, JINI, ...), and

programming languages (Java, TCL, C++, ...)

- 20 - 3 Los - - Alamos = r - - - - - ~ - _ _ _ I - -